Special issue | Virtualized Environments, from Connectivity to Services

Vol. 72, n° 7-8, July-August 2017
Content available on Springerlink 


Guest editors

Emmanuel Bertin, Orange Labs, France
Markus Hofmann, Nokia Bell Labs, USA
Payam Barnaghi, University of Surrey, UK


Virtualized Environments, from Connectivity to Services

Emmanuel Bertin, Markus Hofmann, Payam Barnaghi


Cloud RAN challenges and solutions

Rajeev Agrawal1, Anand Bedekar1, Troels Kolding2, Vishnu Ram3

(1) Arlington Heights, IL, USA
(2) Aalborg, Denmark
(3) Bengaluru, Karnataka, India

Abstract In this paper, we take an overall look at key technical challenges in the evolution of the radio access network (RAN) architecture towards Cloud RAN and solutions to overcome them. To address fronthaul limitations, we examine the implications and tradeoffs enabled by functional splits on fronthaul needs, system performance, and centralization scale. We examine the architecture of algorithms for multi-cell coordination and implications in a Cloud RAN environment. To maximize the use of general-purpose processors (GPP) and operating systems such as Linux for Cloud RAN, we propose methods of achieving real-time performance suitable for RAN functions. To enable right-sizing the amount of compute used for various RAN functions based on the workload, we propose methods of pooling and elastic scaling for RAN functions that exploit the fact that certain RAN functions perform per-user operations while others perform per-cell operations. Cloud RAN also aims to use cloud management technologies such as virtualized infrastructure management (VIM) and orchestration for automating the instantiation and scaling of RAN functions. We identify special needs for RAN arising from real-time constraints and a mix of GPP and non-GPP hardware. In the evolution towards 5G, we propose the use of Cloud-RAN-based multi-connectivity anchoring to address processing bottlenecks in a scalable manner. We identify opportunities for optimization across RAN and other network layers enabled by the distributed edge cloud architecture.

Keywords Cloud RAN, Fronthaul, Real-time, Pooling, Elastic scaling, Multi-cell coordination, Orchestration, NFV, 5G

Penalty migration as a performance signaling method in energy-efficient clouds

Marat Zhanikeev

School of Management, Tokyo University of Science, Fujimi, Japan

Abstract The concept of energy efficiency in clouds has gradually evolved and now includes not only traditional CPU and memory utilizations but also network traffic, number of migrations, etc. Recent literature also shows that the most energy efficient migration schedule is one that, in addition to minimizing migration count, also manages to vacate the most physical machines, thus allowing them to be powered down. While it is obvious that such migration schedules depend on frequent changes in VM-to-PM mapping layouts, existing literature does not offer any practical strategies to that effect. This paper retains the basic notion that lowering the number of migrations and powering down PMs contribute to a level of energy efficiency, but also proposes a new notion called penalty migration. In this paper, the Service and the Cloud have an SLA that spells out an effective range of operations for the Service’s applications, which, if the range is exceeded, can be penalized by the Cloud in several ways discussed in this paper. In addition to higher fluidity, the proposed strategies also improve QoS from the viewpoint of the Cloud’s resources as well as individual service populations. In turn, services can benefit from penalty migrations, treating migration events as QoS monitoring data. This proposal is the first known attempt to distribute the task of performance management between cloud and service providers, where penalty migrations are a form of signaling between the two roles.

Keywords Energy-efficient migration, VM migration, Live migration, Green clouds, Population management, Performance optimization, Cloud resource economy

Consensus-based resource allocation among objects in the internet of things

Virginia Pilloni1, Luigi Atzori1

(1) DIEE, University of Cagliari, Italy

Abstract The pervasive use of smart objects is encouraging the development of the Internet of Things (IoT) vision, where even the most common and simple object is expected to acquire information from the surrounding ambient and to cooperate with other objects to achieve a common goal . In such a heterogeneous and complex scenario, optimal allocation of resources to application tasks (e.g., available energy, computing speed, storage capacity) is paramount to fairly distribute them and not overload some objects. In this paper, we focus on finding the optimal assignment to the physical devices that can perform the same task needed by the running applications. To this, we rely on the technologies that have been already developed around the notion of Virtual Object (VO), which is the digital counterpart of the physical object and is used to augment its functionalities with the use of virtualization technologies. Our contribution is twofold. Firstly, we extend the current functionalities of VOs to make them capable of implementing a distributed strategy for the allocation of tasks among objects: the information model is enhanced to include the Quality of Information (QoI) notion and the possible different architectural solutions are presented. Secondly, we propose a distributed algorithm where VOs negotiate to reach a consensus on resources allocation, in order to distribute the workload among the objects that can cooperate to the same task and to ensure that the QoI requirements are fulfilled. Simulation results show that, compared to a static frequency allocation, the algorithm enhances the performance of the system with an average improvement of 27% in network lifetime and confirms the compliance to QoI requirements.

Keywords Resource allocation, Internet of things, Virtual objects

Subjective perception scoring: psychological interpretation of network usage metrics in order to predict user satisfaction

Jörg Niemöller1, Nina Washington2

(1) Business Area Digital Services, Ericsson, Sweden
(2) Service Systems Research, Ericsson Research, Sweden

Abstract Experiences of users determine their actions and therefore have great influence on the business result of service providers. Thus, it is highly important to know and control these experiences. Decisions and actions are most effective if they are based on individual users and their detailed experience history. Currently available methods for measuring user experience fail in providing sufficient detail or continuous availability. This paper introduces a big data analytics method, which predicts every user’s satisfaction with the service provider. This is performed using an analytics algorithm, adding psychological interpretation to the measured and individually perceived user’s quality of experience. The result is a score that enables individualized marketing and allows understanding the cause of dissatisfaction. It is also able to assign a subjective quality score to network assets.

Keywords Data analytics, User satisfaction, Quality of experience

XMPP-based infrastructure for IoT network management and rapid services and applications development

Enrico Ferrera1, Davide Conzon1, Paolo Brizzi1, Rosaria Rossini1, Claudio Pastrone1, Marc Jentsch2, Peeter Kool3, Carlos Kamienski4, Djamel Sadok5

(1) Istituto Superiore Mario Boella, Torino, Italy
(2) Fraunhofer FIT, Schloss Birlinghoven, St. Augustin, Germany
(3) CNet Svenska AB, Stockholm, Sweden
(4) Universidade Federal do ABC, São Paulo, Brazil
(5) Universidade Federal do Pernambuco (UFPE), Recife, Brazil

Abstract The information technology ecosystem is today facing many radical and methodological changes driven by the Internet-of-Things (IoT): those innovations impact at various levels, ranging from the device-to-device communication paradigms to the value-added services built on top of them. Though several IoT platforms addressing IoT design requirements have recently been raised in State-of-the-Art (SoTA), there is still a lack of platforms and tools which can help end-users to easily develop IoT applications and to configure and manage IoT infrastructures. In order to address these challenges, this work introduces the system development platform (SDP) developed within the IMPReSS project and specifically one of its components, namely, the IoT Platform’s Infrastructure for Configurations (IoT-PIC). It supports developers and users in arranging, configuring, and monitoring the various components of an IoT platform. Specifically, the paper highlights the solution adopted to face two services: IoT Network Management (NM) and platform commissioning. The proposed infrastructure, based on the eXtensible Messaging and Presence Protocol (XMPP), provides means for device discovery and IoT network monitoring, enabling also the mash-up of the platform entities. The platform commissioning tool leverages this feature to compose available modules and services, to implement the desired IoT application. This paper also describes the Resource Adaptation Interface (RAI), which virtualizes physical devices within the IoT platform.

Keywords Internet-of-Things, IMPReSS project, Network Management, Service composition, IoT platform configuration, Model-driven development, XMPP

Cross-domain identity and discovery framework for web calling services

Ibrahim Tariq Javed1, Rebecca Copeland1, Noel Crespi1, Marc Emmelmann2, Ancuta Corici2, Ahmed Bouabdallah3, Tuo Zhang3, Saad El Jaouhari3, Felix Beierle4, Sebastian Göndör4, Axel Küpper4, Kevin Corre5, Jean-Michel Crom5, Frank Oberle6, Ingo Friese6, Ana Caldeira7, Gil Dias7, Nuno Santos7, Ricardo Chaves7, Ricardo Lopes Pereira7

(1) Institut Mines-Telecom, Telecom Sud-Paris, Evry, France
(2) NGNI, Fraunhofer FOKUS, Berlin, Germany
(3) Department SRCD, IMT Atlantique, Cesson-Sévigné, France
(4) Technische Universität Berlin, Telekom Innovation Laboratories, Berlin, Germany
(5) Orange Labs Products & Services, Cesson-Sévigné, France
(6) Telekom Innovation Laboratories, Berlin, Germany
(7) INESC-ID, IST, Universidade de Lisboa, Lisbon, Portugal

Abstract Cross-domain identity management remains a major challenge for potential WebRTC adopters. In order to provide a global web-based communication system, it is critical to locate the destination called party, map the identity to the user device, and provide mutual authentication for both caller and called party. In this paper, we present a novel identity management and user discovery framework that enables callers to search and locate users across service domains. The identity management is decoupled from the used calling service, allowing users to manage their profiles and credentials independently of the applications. The framework is designed to preserve privacy and exploit web technology to gain trust and contact list management.

Keywords WebRTC, Identity management, Trust, Real-time communication, P2P, Directory, Social graph, Registry

Open Topics

Analysis of capacity limits over fading environment with imperfect channel state information for cognitive radio network

Bindu Bharti1, G Singh1

(1) Department of Electronics and Communication Engineering, Jaypee University of Information Technology, Waknaghat, India

Abstract In this paper, we have explored an optimal power allocation scheme for the spectrum sharing with imperfect channel state information between the cognitive/secondary user (CU) and licensed/primary user (PU) in the Rayleigh fading environment. We have analyzed the ergodic capacity of CU link under the combination of peak transmit power and peak/average interference power constraints with or without the primary user interference. In addition to this, the outage capacity with multiple primary users’ interference is also analyzed with the error variance under the joint peak transmit power and peak interference power constraint as well as individual peak interference power constraint. Moreover, the power disbursement is also investigated to achieve the lower limit of ergodic and outage capacity. The minimum mean square channel estimation technique is used for the channel estimation between CU and PU. Further, the convex optimization method is used for the optimal power allocation.

Keywords Cognitive radio, Ergodic capacity, Outage probability, Minimum mean square error, Power consumption

A new synthesis filter for filter bank systems with transmission delay

Lan Wu1, Chenglin Wen1,2

(1) College of Electrical Engineering, Henan University of Technology, Zhengzhou, China
(2) Institute of Information and Control, Hangzhou Dianzi University, Hangzhou, China

Abstract In this paper, a new synthesis filter is designed for filter bank systems with transmission delay by adapting random time delay or data-missing situations in the communication network. Firstly, we formulate a mathematical description of the data transmission delay and a measurement model. Secondly, the state model and the measurement model are adapted into new forms by expanding their dimensions. In this way, non-white noise can be converted to white noise, which can be dealt with using standard Kalman filtering. Thirdly, a dimensionality reduction method is proposed to overcome the complexity generated by expanding dimensions, and then a synthesis filter is designed with dimensions equal to those of the original system. Finally, a simulation is presented to demonstrate the effectiveness and the advantages of the proposed method.

Keywords Filter bank, Transmission delay, Non-white noise, Synthesis filter, Kalman filter

A trust-based security environment in MANET: definition and performance evaluation

Aida Ben Chehida Douss1, Ryma Abassi1,, Sihem Guemara El Fatmi1

(1) Higher School of Communication, Sup’Com University of Carthage Tunis, Carthage, Tunisia

Abstract Due to the lack of centralized administration or fixed network infrastructure, classic routing protocols cannot be applied in mobile ad hoc network (MANET). Hence, some specific ones have been proposed. Paradoxically, due to its importance in the MANET deployment, routing constitutes a privileged target of attackers. In addition, securing MANET constitutes a hard and challenging task that has attracted many researchers. For our concern, we proposed, in a previous work, a reputation-based trust management scheme built upon a recently proposed mobility-based clustering approach (MCA) which classifies nodes into clusters with one-hop members and elected cluster heads (CHs) and which maintains the organization of the network in the presence of mobility. The whole proposition, baptized trust-based MCA (TMCA), detects and isolates malicious behaviors. In order to improve network performance and to maintain its stability by avoiding the re-invocation of the clustering approach, we propose, in the first step of this work, a delegation process based on TMCA, baptized delegation TMCA-based (DTMCA). This delegation allows the CH to transfer its privileges to a chosen member in case of displacement or energy dissipation. In the second step, simulation experiments are conducted to evaluate the performance of the proposed DTMCA scheme. Finally, in the third step, we suggest to model the entire security process with its three components, namely MCA, TMCA, and DTMCA, in a layered manner. This is realized in order to show the different interactions between these three components.

Keywords MANET, Clustering, Security, Trust management, Delegation

Channel modeling for high-speed indoor powerline communication systems: the lattice approach

Modisa Mosalaosi1, Thomas Afullo1

(1) School of Electrical, Electronic & Computer Engineering, University of KwaZulu-Natal, Durban, South Africa

Abstract The transmission of high-frequency signals over powerlines, known as powerline communications (PLC), plays an important role in contributing toward global goals for broadband services inside the home and office. In this paper, we aim to contribute to this ideal by presenting a powerline channel modeling approach which describes a powerline network as a lattice structure. In a lattice structure, a signal propagates from one end into a network of boundaries (branches) through numerous paths characterized by different reflection/transmission properties. Due to theoretically infinite number of reflections likely to be experienced by a propagating wave, we determine the optimum number of paths required for meaningful contribution toward the overall signal level at the receiver. The propagation parameters are obtained through measurements and other model parameters are derived from deterministic powerline networks. It is observed that the notch positions in the transfer characteristics are associated with the branch lengths in the network. Short branches will result in fewer notches in a fixed bandwidth as compared to longer branches. Generally, the channel attenuation increase with network size in terms of number of branches. The proposed model compares well with experimental data.

Keywords Powerline communications, PLC channel modeling, Lattice structure, Channel attenuation